{"url":"/method/dpn","slug":"dpn","name":"DPN","full_name":"Dual Path Network","full_name_withheld":false,"description_markdown":"A **Dual Path Network (DPN)** is a convolutional neural network which presents a new topology of connection paths internally. The intuition is that [ResNets](https://paperswithcode.com/method/resnet) enables feature re-usage while [DenseNet](https://paperswithcode.com/method/densenet) enables new feature exploration, and both are important for learning good representations. To enjoy the benefits from both path topologies, Dual Path Networks share common features while maintaining the flexibility to explore new features through dual path architectures. \r\n\r\nWe formulate such a dual path architecture as follows:\r\n\r\n$$x^{k} = \\sum\\limits\\_{t=1}^{k-1} f\\_t^{k}(h^t) \\text{,}  $$\r\n\r\n$$\r\ny^{k} = \\sum\\limits\\_{t=1}^{k-1} v\\_t(h^t) = y^{k-1} + \\phi^{k-1}(y^{k-1}) \\text{,} \\\\\\\\\r\n$$\r\n\r\n$$\r\nr^{k} = x^{k} + y^{k} \\text{,} \\\\\\\\\r\n$$\r\n\r\n$$\r\nh^k = g^k \\left( r^{k} \\right) \\text{,}\r\n$$\r\n\r\nwhere $x^{k}$ and $y^{k}$ denote the extracted information at $k$-th step from individual path, $v_t(\\cdot)$ is a feature learning function as $f_t^k(\\cdot)$. The first equation refers to the densely connected path that enables exploring new features. The second equation refers to the residual path that enables common features re-usage. The third equation defines the dual path that integrates them and feeds them to the last transformation function in the last equation.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Dual Path Networks","paper":"/paper/dual-path-networks","first_author":"Yunpeng Chen","n_authors":6,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/dual-path-networks"},"source":{"url":"http://arxiv.org/abs/1707.01629v2","title":"Dual Path Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/osmr/imgclsmob/blob/c03fa67de3c9e454e9b6d35fe9cbb6b15c28fda7/pytorch/pytorchcv/models/dpn.py#L322","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Convolutional Neural Networks","url":"/methods/category/convolutional-neural-networks","pwc_aliases":[]}],"n_papers_tagged":19,"archive_num_papers":19,"papers_newest_first":[{"paper":null,"title":"DPN-GAN: Inducing Periodic Activations in Generative Adversarial Networks for High-Fidelity Audio Synthesis","date":"2025-05-14","arxiv_id":"2505.09091","n_code_links":0,"syntology":null},{"paper":null,"title":"Dynamic Pyramid Network for Efficient Multimodal Large Language Model","date":"2025-03-26","arxiv_id":"2503.20322","n_code_links":0,"syntology":null},{"paper":null,"title":"rECGnition_v2.0: Self-Attentive Canonical Fusion of ECG and Patient Data using deep learning for effective Cardiac Diagnostics","date":"2025-02-22","arxiv_id":"2502.16255","n_code_links":0,"syntology":null},{"paper":null,"title":"Deep Pattern Network for Click-Through Rate Prediction","date":"2024-04-17","arxiv_id":"2404.11456","n_code_links":0,"syntology":null},{"paper":"/paper/generalized-category-discovery-with-decoupled","title":"Generalized Category Discovery with Decoupled Prototypical Network","date":"2022-11-28","arxiv_id":"2211.15115","n_code_links":2,"syntology":null},{"paper":null,"title":"Breaking the Curse of Dimensionality in Multiagent State Space: A Unified Agent Permutation Framework","date":"2022-03-10","arxiv_id":"2203.05285","n_code_links":0,"syntology":null},{"paper":"/paper/mag-a-simple-learning-based-patient-level","title":"MAg: a simple learning-based patient-level aggregation method for detecting microsatellite instability from whole-slide images","date":"2022-01-13","arxiv_id":"2201.04769","n_code_links":1,"syntology":null},{"paper":"/paper/signal-processing-based-deep-learning-for","title":"Signal Processing Based Deep Learning for Blind Symbol Decoding and Modulation Classification","date":"2021-06-19","arxiv_id":"2106.10543","n_code_links":1,"syntology":null},{"paper":"/paper/dpn-senet-a-self-attention-mechanism-neural","title":"DPN-SENet:A self-attention mechanism neural network for detection and diagnosis of COVID-19 from chest x-ray images","date":"2021-05-20","arxiv_id":"2105.09683","n_code_links":1,"syntology":null},{"paper":null,"title":"RL-CSDia: Representation Learning of Computer Science Diagrams","date":"2021-03-10","arxiv_id":"2103.05900","n_code_links":0,"syntology":null},{"paper":"/paper/towards-maximizing-the-representation-gap","title":"Towards Maximizing the Representation Gap between In-Domain & Out-of-Distribution Examples","date":"2020-10-20","arxiv_id":"2010.10474","n_code_links":1,"syntology":null},{"paper":"/paper/dpn-detail-preserving-network-with-high","title":"DPN: Detail-Preserving Network with High Resolution Representation for Efficient Segmentation of Retinal Vessels","date":"2020-09-25","arxiv_id":"2009.12053","n_code_links":2,"syntology":null},{"paper":"/paper/comprehensive-comparison-of-deep-learning","title":"Comprehensive Comparison of Deep Learning Models for Lung and COVID-19 Lesion Segmentation in CT scans","date":"2020-09-10","arxiv_id":"2009.06412","n_code_links":1,"syntology":null},{"paper":null,"title":"Shape Detection In 2D Ultrasound Images","date":"2019-11-22","arxiv_id":"1911.09863","n_code_links":0,"syntology":null},{"paper":"/paper/graph-based-global-reasoning-networks","title":"Graph-Based Global Reasoning Networks","date":"2018-11-30","arxiv_id":"1811.12814","n_code_links":9,"syntology":{"ran":8,"of":15,"unverified":7,"pointer_only":4}},{"paper":null,"title":"Multi-function Convolutional Neural Networks for Improving Image Classification Performance","date":"2018-05-30","arxiv_id":"1805.11788","n_code_links":0,"syntology":null},{"paper":"/paper/deeplung-deep-3d-dual-path-nets-for-automated","title":"DeepLung: Deep 3D Dual Path Nets for Automated Pulmonary Nodule Detection and Classification","date":"2018-01-25","arxiv_id":"1801.09555","n_code_links":2,"syntology":null},{"paper":"/paper/an-analysis-of-scale-invariance-in-object-1","title":"An Analysis of Scale Invariance in Object Detection - SNIP","date":"2017-11-22","arxiv_id":"1711.08189","n_code_links":0,"syntology":null},{"paper":"/paper/dual-path-networks","title":"Dual Path Networks","date":"2017-07-06","arxiv_id":"1707.01629","n_code_links":18,"syntology":{"ran":2,"of":2,"unverified":0,"pointer_only":2}}],"papers_shown":19,"tasks":[{"task":"/task/image-classification","name":"Image Classification","papers":3},{"task":"/task/image-classification","name":"image-classification","papers":3},{"task":"/task/classification-1","name":"Classification","papers":2},{"task":"/task/data-augmentation","name":"Data Augmentation","papers":2},{"task":"/task/decoder","name":"Decoder","papers":2},{"task":"/task/classification","name":"General Classification","papers":2},{"task":"/task/segmentation","name":"Segmentation","papers":2},{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":2},{"task":"/task/action-classification","name":"Action Classification","papers":1},{"task":"/task/action-recognition-in-videos","name":"Action Recognition","papers":1},{"task":"/task/anomaly-detection","name":"Anomaly Detection","papers":1},{"task":"/task/audio-generation","name":"Audio Generation","papers":1},{"task":"/task/audio-synthesis","name":"Audio Synthesis","papers":1},{"task":"/task/click-through-rate-prediction","name":"Click-Through Rate Prediction","papers":1},{"task":"/task/computed-tomography-ct","name":"Computed Tomography (CT)","papers":1},{"task":"/task/diagnostic","name":"Diagnostic","papers":1},{"task":null,"name":"GPU","papers":1},{"task":"/task/image-segmentation","name":"Image Segmentation","papers":1},{"task":"/task/language-modeling","name":"Language Modeling","papers":1},{"task":"/task/language-modelling","name":"Language Modelling","papers":1}],"tasks_shown":20,"n_tasks":47,"usage_by_year":[{"year":"2017","papers":2},{"year":"2018","papers":3},{"year":"2019","papers":1},{"year":"2020","papers":3},{"year":"2021","papers":3},{"year":"2022","papers":3},{"year":"2024","papers":1},{"year":"2025","papers":3}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/dpn"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}